Luke A.J. O'Neill is in the Cytokine Research Group, Department of Biochemistry, Trinity College, Dublin, Ireland. laoneill@tcd.ie
Members of the Toll-like receptor−interleukin 1 receptor superfamily signal inflammatory responses. However, a member of this family is now shown to modulate these responses by acting as a negative regulator.
The initial phase of host defense against invading microbes involves a family of proteins called Toll-like receptors (TLRs). These proteins are expressed on various cell types, most notably dendritic cells, where they act as primary sensors of microbial products and activate signaling pathways that lead to the induction of immune and inflammatory genes. TLRs belong to a broader family of proteins, which include receptors for the pro-inflammatory cytokines interleukin 1 (IL-1) and IL-18 (ref. 1). Among the best-characterized TLRs are TLR4, TLR5 and TLR9, which sense lipopolysaccharide (LPS), flagellin and CpG motifs, respectively. Although these receptors have important functions in host defense, their unrestrained stimulation may be detrimental to the host. Thus, negative regulators of IL-1 receptor (IL-1R), IL-18R and TLRs may be required to modulate their responses. In this issue of Nature Immunology, Wald et al.2 describe an intriguing inhibitor of this receptor superfamily.
All members of the TLR−IL-1R superfamily signal inflammation in a very similar way. This is because they all contain a conserved protein sequence in their cytosolic regions, called the Toll−IL-1R (TIR) domain, which activates common signaling pathways, most notably those leading to the activation of the transcription factor NF-B and stress-activated protein kinases. However, Wald et al. show that an orphan receptor, which has the rather cumbersome but accurate name, single immunoglobulin IL-1R−related protein (SIGIRR)3, is an inhibitory member of this receptor superfamily. SIGIRR seems to temper cellular activation by IL-1, LPS and probably other activators of receptors in the TLR−IL-1R superfamily, such that the biological outcome will be the result of a balance between activation by a receptor and dampening by SIGIRR. SIGIRR therefore acts as a 'brake' on the TLR system, which may be essential for regulating the detrimental effects of innate immunity, as occurs in sepsis and chronic inflammation.
The TLR−IL-1R superfamily can be divided into three subgroups1. The first contains extracellular immunoglobulin (Ig) domains and includes IL-1RI. The second is the TLRs, which lack Ig domains, but have extracellular leucine-rich repeats; recent years have seen tremendous progress in determining their function. The third subgroup consists of upstream adapter molecules, including MyD88, MyD88 adapter-like (Mal, also known as TIRAP) and TIR domain−containing adaptor−inducing interferon- (IFN-; TRIF, also known as TICAM-1). These adapters are recruited to receptor TIR domains and initiate signalling processes through IL-1R−associated kinases (of which there are four) and the adaptor molecule TRAF-6, which leads to activation of four protein kinase cascades, culminating in the activation of NF-B and kinases p38, JNK and p42/p44 MAP kinase1. These molecules in turn promote the production of many proinflammatory proteins and enhance immune reactivity. Recent evidence indicates differences in adapter usage, such that although almost all the receptors recruit MyD88, only some receptors use Mal and TRIF, leading to specificity in outcome4. The best example is TRIF usage by TLR3 and TLR4, which leads to the activation of IFN-regulatory factor 3 and the induction of IFN-5,
6. To some extent, because there has been much more progress made in the understanding of the TLR and adapter subgroups, the Ig subgroup has been neglected. Most members in this subgroup remain as orphan receptors of unknown function, the exceptions being IL-1R and IL-18R and their respective accessory proteins, and IL-1Rrp2, which may be a receptor for IL-1F9, a 'paralog' of IL-1 (ref. 7). Five other IL-1 paralogs occur in humans and it seems likely that they will be ligands or antagonists for the orphan receptors8. Defining a function for SIGIRR therefore assigns a function to one of the orphans; it is of considerable interest that this function is inhibitory for other members of the protein superfamily.
Wald et al. show that SIGIRR is expressed in various mouse tissues, including epithelial cells in the kidney, and is highly expressed in the colon but less so in spleen cells. Bone marrow−derived macrophages did not express SIGIRR. Because LPS down-regulates SIGIRR expression in epithelial cells, this indicates that SIGIRR might be inhibitory, consistent with the fact that SIGIRR has a TIR domain that lacks two amino acids essential for signaling by IL-1RI. To test this possibility, the authors over-expressed SIGIRR in Jurkat and HepG2 cells and demonstrated that it inhibited NF-B activation by IL-1 and IL-18. Clear indications for an inhibitory function came when they constructed SIGIRR- deficient mice, which were shown to be hyper-responsive to LPS and IL-1 but not TNF, whose receptor is not in the TLR−IL-1R superfamily. Primary kidney epithelial cells and splenocytes prepared from these mice were also hyper-responsive to IL-1 and LPS in terms of NF-B and JNK activation. SIGIRR-deficient splenocytes were shown to be hyper-responsive to CpG DNA, in terms of proliferation, indicating the inhibitory effects of SIGIRR may be broader than its effects on IL-1R and TLR4. No difference was found between SIGIRR-deficient and wild-type macrophages, consistent with the lack of expression of SIGIRR in these cells. Finally, the authors show that SIGIRR is associated with IL-1RI, TLR4, TLR5, TLR9 and TRAF-6. IL-1 treatment promoted the association of SIGIRR with IL-1RI and also led to the formation of a complex between SIGIRR, IL-1R-associated kinase (IRAK) and TRAF-6. These mechanistic studies indicate that SIGIRR works by being recruited through its TIR domain to the TIR domain of target receptors, where it might sequester the key signalling proteins IRAK and TRAF-6, and prevent signal propagation (Fig. 1). The evidence for this mechanism is, however, circumstantial and it is possible that SIGIRR launches a negative signal.
Figure 1. The 'on' and 'off' of TLR4 signal transduction.
(a) TLR4 signal transduction is initiated by the recruitment of three TIR domain-containing adapters: MyD88, Mal and TRIF. IRAK-4 is then recruited to the complex, where it phosphorylates IRAK-1, leading to TRAF-6 engagement and the activation of the kinase TAK-1, in a complex with TAB2. IRAK-2 and Tollip are also involved in TLR4 signal transduction, although their precise involvement is not fully worked out. The transcription factor NF-B and several MAP kinase cascades lie downstream and lead to enhanced transcription and translation of inflammatory and immune genes. (b) Negative regulators of TLR4 signal transduction. SIGIRR is a transmembrane protein with a TIR domain, which is recruited to TLR4 and blocks signaling by sequestering IRAK and TRAF-6. MyD88s is a splice variant of MyD88, which prevents IRAK-4 recruitment. IRAK-M inhibits TLR4 signaling by preventing dissociation of IRAK from the signaling complex. It is likely that a balance between activation and inhibition of signaling is responsible for the output from TLR4 and other members of the TLR−IL-1R family such as IL-1RI, which show similar negative controls.
SIGIRR seems to act as a pan-inhibitor of the receptor superfamily, although further evidence in support of this notion is needed, in particular with regard to TLR9 and TLR5. The other known endogenous inhibitors of TLR signalling are MyD88s, IRAK-M and SOCS-1 (refs. 9,10,11). MyD88s, which is a splice variant of MyD88, prevents IRAK-4 recruitment9. IRAK-M also interferes with IRAK function by preventing dissociation of IRAK from the signalling complex10. The mechanism of action of SOCS1 is not known. Mice deficient in IRAK-M or SOCS1 have a very similar phenotype to that of SIGIRR-deficient mice in terms of LPS hyper-responsiveness, and both IRAK-M and SOCS-1 are important in LPS tolerance, in which cells become unresponsive to LPS. SIGIRR may therefore also be involved in this process, particularly in epithelial cells. SIGIRR may in fact be important for tolerance to microbes in the gut and other epithelial tissues.
The findings of Wald et al. are of interest for the development of inhibitors of TLR function, which could be used to treat sepsis and inflammatory diseases. The proposed interaction between the TIR domain of SIGIRR and the TIR domains of target receptors indicates that it might be possible to interfere with the TIR domain function with inhibitors. An inhibitor based on the so-called BB loop in the TIR domain has been developed12; it seems to act by preventing recruitment of adapters to receptor TIR domains. Perhaps SIGIRR has a similar obscuring effect. Future studies will need to determine the mechanism of SIGIRR recruitment and to define what it is about its TIR domain that makes it inhibitory. A final issue regarding mechanism concerns a ligand for SIGIRR. Even though it only has one Ig domain, SIGIRR may still have a ligand, which could be one of the IL-1 paralogs. No ligand was used in the study by Wald et al., however, although over-expression of SIGIRR might have effects similar to ligand engagement. Like other receptor systems in immunity, the TLR−IL-1R superfamily now has an inhibitory member. Further analysis will continue to improve our understanding of the functions of this fascinating and essential set of receptors in innate immunity and inflammation.
B and stress-activated protein kinases. However, Wald et al. show that an orphan receptor, which has the rather cumbersome but accurate name, single immunoglobulin IL-1R−related protein (SIGIRR)
(IFN-
